Re: NCRP bias?

[BERNARD L COHEN <blc+@PITT.EDU>]

On Mon, 5 May 2003, Carl Miller wrote:



>

> Also, I think it has been pointed out on this list, before that Dr. Cohen

 fails to consider the interaction between all the colinear factors that can

 work together such as smoking, education, and socioeconomic status at one time.

  These factors are very inter related and Dr. Cohen looks at his potentially

 confounding factors in a univariate manner rather than a multivariable manner

 which would be required to capture the effect.



	--This specific problem is addressed in Section 3.3 of Item #7 on

my web site. In fact your questions are more broadly answered in all of

Section 3 of that paper

> Dr. Cohen may not think it is plausible to explain his findings just by his use of guestimated county smoking rates(which are inversely related to the radon), but what about the co-linearity between smoking, education and socioeconomic factors? Dr. Field has pointed this out before.

>

> In addition, I think it was Drs. Lubin and Field that showed when better data

 was for lung cancer incidence the inverse assocaition was not found in Iowa

 (a s tate with high radon).  Dr. Cohen has never explained this finding in a

 satisfactory manner.



	--Statistics for the entire U.S. are far more robust than

statistics for a single state. There are a few states where the slope of

the regression of lung cancer on radon levels is not negative, but for the

great majority of states it is negative, and the average of slopes for all

states is negative and close to that obtained for the entire U.S. This

difference among different states is easily understood as due to

statistics. See Table 4 of Item #2 on my web site.



************************************************************************

You are currently subscribed to the Radsafe mailing list. To unsubscribe,

send an e-mail to Majordomo@list.vanderbilt.edu  Put the text "unsubscribe

radsafe" (no quote marks) in the body of the e-mail, with no subject line.

You can view the Radsafe archives at http://www.vanderbilt.edu/radsafe/